Finally, the retrieval algorithm was applied to BrO ACDs, with the results supplied in Table 6.10. Unfortunately, there are no other sources of BrO measurements for direct comparison. However, an in-situ BrO instrument flew on the ER-2 for the ASHOE/MAESA campaign of 1994. Mid-latitude and tropical BrO mixing ratios varied between 4-10 pptv () near 20 km. Over the Antarctic during the southern hemisphere spring, mixing ratios varied from 8-13 pptv. The VCD in each of the three layers from Table 6.10 has been converted into a mixing ratio by assuming each layer to be well mixed, also given in Table 6.10. This will allow for a rough comparison with the limited number of published BrO measurements. The retrievals were performed by constraining the VCD below the ER-2 to the value obtained from the nadir ACD after applying the appropriate air mass factor (see section 6.6.2). The VCD above the ER-2 was constrained to 2.5 cm-2.
Of the two scans previously examined from 6 May, only the first one (2048 UTC) produced meaningful results. Mixing ratios of 5.6, 5.7 and 14.2 pptv were retrieved in C1, C2, and C3, respectively. A retrieval was performed on the scan immediately following this one and resulted in near identical mixing ratios. The spectra making up the second scan (2248 UTC) produced poor DOAS fits although the ACDs were not below the detection limit. The situation was similar throughout the 21 September 1997 flight as the majority of these fits were also quite poor. In fact, during portions of this flight small negative ACDs were fit. It is currently unknown why this occurred.
a 8 June 1997
|
In lieu of the 21 September flight, the flight made on 7 June is analysed. During this flight, the ER-2 proceeded due north to 90N before retracing its path back to Fairbanks. This flight produced BrO DOAS fits which were quite good throughout the return (southbound) leg but not nearly as good throughout the northern leg although it is interesting to note that the ACDs were similar. Retrievals have been performed for two scans from this flight. The first gave values of 7.5 pptv and 6.7 pptv in C3 and C2, respectively, while the second yielded 6.2 pptv and 14.0 pptv. For the second scan, in particular, the ACDs seemed to peak for lower elevation angles than was observed for the previous scans and this forced more of the BrO into the middle layer. The 14.0 pptv in this layer does not seem realistic, based on the modelled behaviour of BrO with altitude (see below).
The rapid increase in BrO with height throughout the lower stratosphere
as observed on 6 May and 7 June (first scan)
is generally consistent with the limited
number of published BrO measurements,
including those of Brune et al. (1988).
Although the vertical resolution of these profiles is very coarse,
the 6 May measurements
are similar to those made during perturbed conditions from 16-22 km
(McKinney et al., 1997), which should be roughly applicable given the time
of year and latitude of this scan.
The 7 July measurements (first scan) is comparable
with results from the Canadian
Middle Atmosphere Model (CMAM) (de Grandpré et al., 1997) and
one-dimensional model results (Fish et al., 1995) for unperturbed
conditions. A more detailed comparison with chemical models can
only be carried out when the same time of year and day are used.
This increase with height is the result of a combination
of an increase in total inorganic bromine, Bry
(
), with height as well as a repartitioning
among the species comprising Bry.
In the lower stratosphere, the BrO/Bry ratio is controlled mainly by
the reactions between Br, BrO, and BrONO2. As ozone increases with
height, BrO will also tend to increase through the reaction,
(11.36) |
(11.37) |
Retrievals were also performed for the 26 April flight. The VCDs below the ER-2 from this flight were larger than previous flights and it is believed that most of this resides in the troposphere (see section 6.6.2). The retrieval placed nearly all of this column in C1, as expected, but there were increased amounts in the stratosphere also. The 12-16 km layer had approximately twice that observed from the 6 May flight while the 16- km layer was about 50% larger. The ACDs from this flight were about 50% larger than those from the other flights. While the limb steps are not very sensitive to the tropospheric air, such a large absorption signal in the troposphere must have an impact on the retrieved limb ACDs as some surface reflected light will reach the ER-2. While it is possible that these larger amounts of BrO in the stratosphere are real and perhaps even related to the increased nadir-ACDs below the aircraft, it seems more likely that this is an artifact of the retrieval, especially as the fractional increase in C2 is larger than C3 (as compared with the 6 May results).